Polymer Degradation and Stability 83 (2004) 281–287

www.elsevier.com/locate/polydegstab

Direct observation of polyhydroxyalkanoate

granule-associated-proteins on native granules and on
poly(3-hydroxybutyrate) single crystals by atomic force microscopy
Kumar Sudesha,*, Akira Maeharab, Zhihua Ganb, Tadahisa Iwatab, Yoshiharu Doib,c
a
School of Biological Sciences, Universiti Sains Malaysia 11800, Penang, Malaysia
b
Polymer Chemistry Laboratory, RIKEN Institute, Hirosawa 2-1, Wako-shi, Saitama 351-0198, Japan
c
Department of Innovative and Engineered Materials, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226-8502, Japan

Received 22 April 2003; accepted 5 June 2003

Abstract
Polyhydroxyalkanoate (PHA) granules isolated from Comamonas acidovorans revealed the presence of globular particles on the
granule surface when imaged using atomic force microscopy (AFM). The globular particles appeared to form a monolayer on the
granule surface. Repeated washing with pure water resulted in granules with smooth surface without any globular particles. Purified
granule-associated-proteins and PHA single crystals were used to construct a model representing native granule surface for AFM
imaging. For this purpose, PhaR and PhaP proteins, purified from Paracoccus denitrificans were adsorbed onto poly(3-hydro-
xybutyrate) [P(3HB)] single crystals. PhaR is a 22 kDa protein involved in the regulation of PhaP (phasin) protein (16 kDa). Both
purified proteins were morphologically different but adsorbed uniformly on the single crystal surface forming a monolayer.
Immunogold-labeling was used to further confirm the adsorption and identity of PhaR protein on P(3HB) single crystals. The study
shows for the first time, that proteins associated with PHA granules can be imaged directly and characterised using AFM. In
addition, AFM images obtained in this study provide direct evidence for the binding of PhaR and PhaP proteins to the hydro-
phobic PHA surface.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Polyhydroxyalkanoate (PHA) granules; Atomic force microscopy (AFM)

1. Introduction In a previous study we have shown that AFM offers

a novel approach to study the surface structure of
Atomic force microscopy (AFM) [1] offers a novel polyhydroxyalkanoate (PHA) granules [6]. Earlier
method of imaging for biologists to study the surface attempts to elucidate the native surface structure of
architecture of cells and cellular components. The abil- PHA granules using conventional microscopic techni-
ity of AFM to provide real-time three-dimensional ques have proven to be inconclusive. This is because
images under natural conditions with molecular resolu- preparation of the PHA granules for characterisation
tion has been used to obtain images of various samples using electron microscopy usually involves staining and
[2–4]. Besides providing high resolution images of sur- dehydration. These treatments, especially dehydration,
face topography, AFM can also be used to obtain force can greatly affect the native structure of PHA granules
measurements to probe the physical properties of sam- (author’s unpublished observation).
ples, such as molecular interactions and mechanical PHAs are high molecular weight carbon and energy
properties [5]. reserve materials that exist in the form of water-inso-
luble granules in the cell cytoplasm of many micro-
organisms. Although most PHAs exhibit high
* Corresponding author. Tel.: +60-4-6533888; fax: +60-4-
crytallinity when extracted from the cells, in vivo the
6565125. granules are intriguingly maintained in an amorphous
E-mail address: ksudesh@usm.my (K. Sudesh). state [7,8]. Besides the enzymes involved in the bio-
0141-3910/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0141-3910(03)00273-8
282 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287

synthesis (PHA synthase) and mobilization of PHAs described in our recent publication [6]. Briefly, the
(intracellular PHA depolymerases), the native granules PHA granules used in this study was isolated from
are associated with several kinds of proteins termed Comamonas acidovorans (JCM 10181) that was culti-
phasins (PhaP), that are important in the formation and vated in the presence of 1% (wt./vol.) sodium pentano-
stabilisation of the granules [9]. These proteins are ate. Gas chromatography analysis revealed that the
located on the granule surface [10,11]. However, it is not PHA is a copolymer of poly(3-hydroxybutyrate-co-75
known whether the proteins form a continuous layer mol% 3-hydroxyvalerate) [P(3HB-co-75 mol% 3HV)].
around the PHA granules, thus creating a boundary Crude cell lysate of C. acidovorans was prepared by
between the hydrophobic granule and the hydrophilic passage of the harvested cells through a French pressure
cytoplasm. cell twice at 96 MPa. PHA granules were purified by
Interest in the surface architecture of native PHA centrifuging the lysate on a discontinuous sucrose den-
granules has been gaining attention in recent years. sity gradient (210,000 g, 90 min, 4  C). The purified
Although it is known that several phasin species are granules were washed once with 0.1 M Tris–HCl (pH 7.4)
bound to the granule surface their function(s) still before adsorption onto a glass cover slip. The glass cover
remain elusive [12]. Phasins have been shown to influ- slips used in this study were layered with ultrathin poly-
ence the size and number of PHA granules in the cell cationic film for firm adsorption of the PHA granules.
[13–15]. For the binding of these phasins to the PHA
granules, two different models have been proposed 2.2. Preparation of P(3HB) single crystals
based on the phasins amino acid sequence information.
The first model assumes the presence of a phospholipid Detailed description of the P(3HB) single crystal pre-
monolayer to which the phasins are anchored [16,17]. In paration is available elsewhere [23]. The method was
the second model, it was proposed that the phasins are derived from that of Marchessault and coworkers [24].
directly bound to the PHA granules because all of them
contain at least one markedly hydrophobic region [18]. 2.3. Purification of PhaR, PhaP proteins and preparation
Despite the many uncertainties in their actual function, of antibodies
phasins clearly affect PHA biosynthesis and accumula-
tion in the cell [19]. Recent studies have shown that PhaR (22 kDa) protein was purified from recombi-
some phasins are regulated by specific DNA-binding nant Escherichia coli BL21 (DE3) cells overexpressing
proteins that can also bind to the PHA granules [20–22]. the phaR gene while PhaP (16 kDa) protein was pur-
PhaR has been identified as the protein involved in the ified from E. coli XL1 Blue (pPDPK1.7) cells over-
regulation of phasin expression. It has been proposed expressing the phaP gene [15,25]. Antibodies raised
that PhaR derepresses phaP expression by binding to against the PhaR protein were prepared in rabbit as
the PHA granule [20]. PhaR is therefore the first class described previously [15]. Anti-rabbit immunoglobulin
IV granule-associated-protein [9] to be identified and G (IgG) (whole molecule) gold conjugate (10 nm nom-
characterised at the molecular level. inal particle) was purchased from Sigma Chemical
The primary interest of this study is to show that Company (St. Louis, Mo.). Four microlitres of the
AFM can be used to observe directly and characterise purified PhaR (and PhaP) protein (0.2 mg/ml) were
proteins associated with native PHA granules. AFM is mixed with a 1 ml suspension of P(3HB) single crystals
probably the best imaging tool for this purpose because (0.4 mg/ml) in 50 mM Tris–HCl buffer (pH 7.0). Fol-
of the minimum sample pretreatment that is required lowing a 30 min incubation at 37  C to allow the pro-
and also because AFM in the tapping mode is gentle teins to adsorb on to the P(3HB) single crystals, the
enough to avoid sample damage. Evidence has been mixture was washed three times by centrifugation with
obtained to show that the proteins associated with PHA Buffer 1 (0.1 M maleic acid, 0.15 M NaCl, pH 7.5)
granules are capable of forming a continuous mono- supplemented with 1% skim milk and 0.5 M NaCl.
layer on PHA granules and single crystals. The finding in The washed P(3HB) single crystals were then resus-
this study strongly supports the idea that PhaR protein pended in 1.0 ml of the same buffer mixture before
binds directly to the hydrophobic PHA granule surface. adding 10 ml antiserum against PhaR and incubating at
room temperature for 3 h. To remove unbound anti-
serum, the P(3HB) single crystals were washed once
2. Materials and methods with the buffer mixture, followed by twice with Buffer
1 containing only 0.5 M NaCl and finally once with
2.1. Biosynthesis, purification and adsorption of PHA only Buffer 1. The single crystals were then resus-
granules pended in 200 ml Buffer 1 added with 0.1% poly-
ethylene glycol (PEG) (Mw=20,000). Ten microlitres of
Detailed information on the biosynthesis, purifica- anti-rabbit IgG gold conjugate were then added to the
tion and adsorption of the PHA granules have been mixture and incubated at room temperature for 30 min.
 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287 283

The gold labelled preparation was washed once with

Buffer 1 containing 0.05% PEG, once with Buffer 1
containing 0.02% PEG, and finally resuspended in
Buffer 1 containing 0.02% PEG for deposition on the
silicon substrate.

2.4. Atomic force microscopy observation

AFM analysis was performed with an SPI3800/

SPA400 (Seiko Instruments Inc.) using the tapping
mode. Pyramid-like silicon tips, made of Si3N4 mounted
on 200 mm long microcantilever with spring constants of
13 Nm 1 were used for the dynamic force mode mea-
surements. Simultaneous registration was performed for
height (topographic) and deflection images. Although
the deflection image could not give any quantitative
height information, it is a useful mode to reveal height
differences in the sample surface, especially when the
sample surface is rough.

3. Results and discussion

3.1. Globular particles on the PHA granule surface

Fig. 1. (A) AFM deflection images of P(3HB-co-3HV) granule iso-
lated from C. acidovorans showing the presence of globular particles
In a previous study, we have tested various substrates on the granule surface. (B) Close up observation of the globular par-
(glass cover slip, glass cover slip coated with poly- ticles in (A). (C) P(3HB-co-3HV) granule that has been subjected to
anionic film and/or polycationic film, silicon, freshly repeated washing with water, showing the absence of globular par-
cleaved mica) to adsorb the freshly isolated native PHA ticles. (D) AFM height image of the globular particles in (B). Inset
granules [26]. The granules have to be firmly adsorbed showing the apparent size of a globular particle in (D).
to a flat substrate in a gentle manner, for scanning by
the AFM tip. PHA granules were successfully imaged PHA granules prior to the onset of crystallisation
on all the tested substrates. However, for repeated, long (manuscript in preparation). The smooth surfaced,
term observation, PHA granules adsorbed on glass washed PHA granule in Fig. 1C eventually developed
cover slips coated with polycationic ultrathin film crystal-like lamellar structures as observed in a previous
proved superior [6]. Fig. 1 shows AFM images of study [6]. It should be noted that the PHA granules are
P(3HB-co-3HV) granule adsorbed on glass cover slip adsorbed very firmly on the polycationic film and can
coated with polycationic film. The granule is char- withstand repeated washing. However, the washing
acterised by the presence of globular particles on its process can sometime result in granule deformation
surface (Fig. 1A). Such globular particles were not which is characterized by an apparent orientation of the
observed on the polyelectrolyte film surface, suggesting adsorbed granules in the direction of the water stream
that the particles bind preferentially to the P(3HB-co- used for washing [6].
3HV) granule. Fig. 1B is a higher magnification AFM Proteins associated with PHA granules are bound
deflection image of the globular particles on the granule. quite firmly to the granule surface. This is based on the
The magnified image shows that the globular particles observation that SDS-PAGE analysis of washed gran-
are adsorbed homogeneously, almost forming a con- ules reveals the firmly adsorbed proteins. In this study,
tinuous layer on the granule surface. we found that the globular particles on the P(3HB-co-
To this end, the adsorbed granules have not been 3HV) granule surface could be removed by repeated
subjected to repeated rinsing with pure water (MiliQ). washing with pure water. The relative ease at which the
The AFM deflection image in Fig. 1C shows the surface globular particles could be removed by gentle washing
structure of a P(3HB-co-3HV) granule that was rinsed maybe because the granules exist singly on the substrate
several times using a gentle stream of pure water. This and not in bulk as is in the preparation for SDS-PAGE
treatment resulted in the disappearance of the globular analysis. In a separate study we have treated freshly
particles, revealing a rather smooth surface structure. isolated PHA granules with aqueous acetone, which
We have obtained evidence to show that this smooth also resulted in the removal of surface particles. In
surface structure is the characteristic of all amorphous addition, acetone treatment also resulted in the swelling
284 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287

of PHA granules and release of polymer microfibrils

[27]. Similar acetone treatment of some PHA granules
[poly(4-hydroxybutyrate)] resulted in granules with very
smooth surface structure initially, but these granules
soon showed high tendency to crystallize (unpublished
observation).
Many of the globular particles on the P(3HB-co-3HV)
granule showed an apparent lateral dimension of about
31 nm across as is shown in the inset of Fig. 1D. Smaller
globular particles could also be observed. Such globular
particles were also observed in a previous study,
although at a much reduced concentration [6]. In the
previous study the adsorbed granules were washed sev-
eral times with pure water, which explains the reduction
in globular particle concentration on the granule sur-
face. At this stage, it is not known if the globular parti-
cles represent individual protein molecules or complexes
consisting of several protein molecules. However, it is
known that the lateral dimension determined by AFM Fig. 2. (A) AFM deflection image of a P(3HB) single crystal on
is subjected to magnification effect caused by tip geo- silicon substrate. (B) Close up deflection image showing the surface
metry [28]. Further AFM observations were therefore structure of P(3HB) single crystal. (C) AFM height image of a
P(3HB) single crystal. Inset showing the height profile of the P(3HB)
conducted using nanostructures with known dimensions
single crystal in (C).
to determine the extent of magnification in this study.

3.2. Globular particles of PhaR proteins on P(3HB)

single crystals

In order to determine further the ability of AFM to

image granule-associated-proteins and to estimate their
sizes, we have used well known nanostructures such as
P(3HB) single crystals as substrates for the adsorption
of purified granule-associated-proteins. In recent years,
AFM has gained popularity as an additional tool to
study the enzymatic degradation of PHAs [23,29].
P(3HB) single crystals have been widely used as a model
substrate for this purpose. Fig. 2A shows the AFM
deflection image of a P(3HB) single crystal on a silicon
substrate. At higher magnification (Fig. 2B), the surface
of the single crystal appeared rather smooth under the
imaging conditions used in this study. Fig. 3C shows the
AFM height image of the P(3HB) single crystal. The
inset gives the height profile at the location indicated by
dotted lines in Fig. 3C. The lamellar thickness of the
Fig. 3. (A) AFM deflection image of PhaR proteins adsorbed on
P(3HB) single crystal used in this study is approximately
P(3HB) single crystals. (B) Close up observation of the globular PhaR
5 nm. The morphology and dimensions of P(3HB) sin- proteins on P(3HB) single crystal. (C) AFM height image of (B). Inset
gle crystals observed in this study are in close agreement showing the height profile of P(3HB) single crystal adsorbed with
with previously reported observation using contact PhaR proteins.
mode AFM [23].
Based on current knowledge of native PHA granule esis and binds to the intracellular PHA granules [20]. In
surface, we attempted to construct a possible model of contrast to the smooth surface of P(3HB) single crystals
the granule surface for AFM observation. For this pur- in Fig. 2, the single crystals are now characterized by a
pose, we adsorbed purified granule-associated-proteins surface covered with globular particles (Fig. 3A). At
onto the P(3HB) single crystals. Fig. 3 shows the AFM higher magnification (Fig. 3B), it can be seen that the
images of P(3HB) single crystals adsorbed with PhaR globular particles are adsorbed homogeneously, almost
protein. PhaR protein has been shown to be a reg- forming a monolayer on the single crystal surface.
ulatory protein that indirectly controls PHA biosynth- Fig. 3C shows the AFM height image of Fig. 3B and its
 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287 285

height profile is shown in the inset. The thickness of the single crystal surface (results not shown). It is interest-
P(3HB) single crystal together with the monolayer of ing to note that the globular particles now seen on the
PhaR proteins is now 12.5 nm. Considering that the P(3HB) single crystals are very regular not only in size
PhaR proteins would adsorb homogeneously on the top but also in shape. Such particles can also be observed
as well as on the bottom surfaces of the P(3HB) single scattered on the silicon substrate albeit at a much
crystals, the PhaR proteins had formed a layer of reduced concentration (Fig. 4A). Higher magnification
approximately 4 nm thick on each side. In a previous of the globular particles on the substrate (Fig. 4C)
study, a 14 kDa PHA granule-associated-protein of shows that each globular particle is identical in size and
Rhodococcus ruber was shown to be a globular protein shape (inset at the bottom). These identical structures
with a diameter of 3 nm using electron microscopy [17]. were interpreted as individual IgG–gold complexes. At
To further confirm that the globular particles on the higher magnification it was clear that the globular
P(3HB) single crystals are indeed PhaR proteins, structures on the P(3HB) single crystal were the same as
immunogold labeling against the PhaR protein was those on the silicon substrate (results not shown).
performed. Fig. 4A and B show the AFM deflection and The average height of the globular particles on the
height images, respectively, of immunogold labeled silicon substrate was about 14 nm (inset at the bottom).
PhaR proteins on P(3HB) single crystal. As expected, This is the combined height of gold particle (10 nm) and
the immunogold labels were detected on the P(3HB) the IgG protein. The IgG protein in this case appears as
single crystals. This was confirmed by transmission the flattened structure around the gold particle (Fig. 4C).
electron microscopy, where it was observed that the The highest point in the IgG complex is interpreted as
gold particles were distributed homogeneously on the the 10 nm gold particle. Colloidal gold particles have
been used as incompressible imaging standards for
AFM [28]. They are known to give constantly accurate
height value when measured by AFM. However, the
lateral dimension of the gold particle is affected by the
AFM tip geometry. The magnification effect of AFM is
a well-known phenomenon that needs to be considered
especially when measuring objects smaller than the tip
size. Using the imaging conditions in this study, the
apparent lateral dimension of the 10 nm gold particles
was approximately 29 nm (inset at the top in Fig. 4).
On the other hand, the apparent lateral dimension of a
single IgG–gold complex was approximately 65 nm
(inset at the bottom in Fig. 4). Repeated scanning of
the IgG–gold complex on the silicon substrate resulted
in structural deformation of the IgG protein but the
gold particle constantly gave similar height values
(results not shown).

3.3. Globular particles of PhaP proteins on P(3HB)

single crystals

We were also able to image another granule-asso-

ciated-protein, PhaP (16 kDa), purified from P. deni-
trificans. Fig. 5 shows the AFM deflection images of
PhaP protein adsorbed on P(3HB) single crystals. The
PhaP protein like PhaR, adsorbed homogeneously on
the single crystals (Fig. 5A). Comparison of the image
in Fig. 5A with that in Fig. 3B, shows that both proteins
are morphologically different. The globular particles of
Fig. 4. (A) AFM deflection image of immunogold-labeled PhaR pro- PhaP protein appear relatively smaller than those of
teins on P(3HB) single crystal. (B) AFM height image of (A). (C) PhaR protein. Both proteins however seem to form a
AFM deflection image of immunogold complex on silicone substrate. continuous monolayer on the P(3HB) single crystal.
(D) AFM height image of (C). Inset (top) showing the apparent lateral Recent studies have shown that phaP expression is
size of gold particles on the complex surface. Inset (bottom) showing
the height profile and apparent lateral size of immunogold complex.
repressed by the binding of PhaR to a region upstream
Gold particle in the complex centre is clearly distinguishable by its of phaP. When P(3HB) granules start to form in the
regular height and shape. cell cytoplasm, PhaR is titrated from phaP, possibly by
286 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287

[2] Ahimou F, Touhami A, Dufrene YF. Real-time imaging of the

surface topography of living yeast cells by atomic force micro-
scopy. Yeast 2003;20(1):25–30.
[3] Hansma HG, Pietrasanta LI, Auerbach ID, Sorenson C, Golan R,
Holden PA. Probing biopolymers with the atomic force micro-
scope: a review. J Biomater Sci Polym Ed 2000;11(7):675–83.
[4] Lal R, John SA. Biological applications of atomic force micro-
scopy. Am J Physiol 1994;266(1 Pt 1):C1–21.
[5] Dufrene YF. Atomic force microscopy, a powerful tool in
microbiology. J Bacteriol 2002;184(19):5205–13.
[6] Sudesh K, Gan Z, Maehara A, Doi Y. Surface structure, mor-
phology and stability of polyhydroxyalkanoate inclusions char-
Fig. 5. (A) AFM deflection image of PhaP proteins adsorbed on acterised by atomic force microscopy. Polym Degrad Stab 2002;
P(3HB) single crystals. (B) Close up image of the globular PhaP pro- 77:77–85.
teins on P(3HB) single crystal. [7] Barnard GN, Sanders JK. The poly-b-hydroxybutyrate granule
in vivo. A new insight based on NMR spectroscopy of whole
cells. J Biol Chem 1989;264(6):3286–91.
binding to the P(3HB) granules [22]. Strong evidence [8] Calvert P. Tender morsels for microbes. Nature 1992;360:535.
for the interaction of PhaR to amorphous as well as [9] Steinbüchel A, Aerts K, Babel W, Föllner C, Liebergesell M,
crystalline P(3HB) granules was also presented [20]. The Madkour MH, Mayer F, Pieper-Fürst U, Pries A, Valentin HE,
present AFM study has provided conclusive evidence Wieczorek R. Considerations on the structure and biochemistry
of bacterial polyhydroxyalkanoic acid inclusions. Can J Micro-
for the direct binding of both PhaR and PhaP proteins biol 1995;41(Suppl. 1):94–105.
to the hydrophobic surface of P(3HB). The results also [10] Gerngross TU, Reilly P, Stubbe J, Sinskey AJ, Peoples OP.
indicate that both PhaR and PhaP interact directly with Immunocytochemical analysis of poly-b-hydroxybutyrate (PHB)
the hydrophobic PHA material without any need for synthase in Alcaligenes eutrophus H16: Localization of the syn-
additional factors in the bacterial cell. thase enzyme at the surface of PHB granules. J Bacteriol 1993;
175(16):5289–93.
[11] Wieczorek R, Pries A, Steinbuchel A, Mayer F. Analysis of a 24-
kilodalton protein associated with the polyhydroxyalkanoic acid
4. Conclusion granules in Alcaligenes eutrophus. J Bacteriol 1995;177(9):2425–
35.
[12] Valentin HE, Stuart ES, Fuller RC, Lenz RW, Dennis D. Inves-
To the best of our knowledge, this is the first study to
tigation of the function of proteins associated to poly-
show that proteins associated with native PHA granules hydroxyalkanoate inclusions in Pseudomonas putida BMO1. J
can be visualised directly using AFM. We were able to Biotechnol 1998;64(2-3):145–57.
obtain three-dimensional images of protein particles on [13] Fukui T, Kichise T, Iwata T, Doi Y. Characterization of 13 kDa
native PHA granules as well as on P(3HB) single crystals. granule-associated protein in Aeromonas caviae and biosynthesis
The results have clearly shown that granule-associated- of polyhydroxyalkanoates with altered molar composition by
recombinant bacteria. Biomacromolecules 2001;2(1):148–53.
proteins are capable of binding directly to the hydro- [14] Jossek R, Reichelt R, Steinbüchel A. In vitro biosynthesis of
phobic PHA material to form a continuous monolayer. poly(3-hydroxybutyric acid) by using purified poly(hydroxyalk-
With the help of immunogold-labeling, it should be pos- anoic acid) synthase of Chromatium vinosum. Appl Microbiol
sible to map the location and distribution of various Biotechnol 1998;49:258–66.
granule-associated-proteins more precisely. Direct visua- [15] Maehara A, Ueda S, Nakano H, Yamane T. Analyses of a poly-
hydroxyalkanoic acid granule-associated 16-kilodalton protein
lisation would enable us to understand in more detail the and its putative regulator in the pha locus of Paracoccus deni-
molecular mechanism of PHA granule formation, mobi- trificans. J Bacteriol 1999;181(9):2914–21.
lization and the regulatory machinery involved. [16] Pieper-Fürst U, Madkour MH, Mayer F, Steinbüchel A. Identi-
fication of the region of a 14-kilodalton protein of Rhodococcus
ruber that is responsible for the binding of this phasin to poly-
hydroxyalkanoic acid granules. J Bacteriol 1995;177(9):2513–23.
Acknowledgements [17] Pieper-Fürst U, Madkour MH, Mayer F, Steinbüchel A. Pur-
ification and characterization of a 14-kilodalton protein that is
This work was supported in part by Ecomolecular bound to the surface of polyhydroxyalkanoic acid granules in
Science Research (RIKEN Institute). KS acknowledges Rhodococcus rubber. J Bacteriol 1994;176(14):4328–37.
financial support from the Fundamental Research [18] Hanley SZ, Pappin DJC, Rahman D, White AJ, Elborough KM,
Slabas AR. Re-evaluation of the primary structure of Ralstonia
Grant Scheme of Universiti Sains Malaysia (304/PBIO- eutropha phasin and implications for polyhydroxyalkanoic acid
LOGI/670038). granule binding. FEBS Lett 1999;447:99–105.
[19] York GM, Stubbe J, Sinskey AJ. New insight into the role of the
PhaP phasin of Ralstonia eutropha in promoting synthesis of
polyhydroxybutyrate. J Bacteriol 2001;183(7):2394–7.
References
[20] Maehara A, Taguchi S, Nishiyama T, Yamane T, Doi Y. A
repressor protein, PhaR, regulates polyhydroxyalkanoate (PHA)
[1] Binnig G, Quate CF, Gerber C. Atomic force microscopy. Phys synthesis via its direct interaction with PHA. J Bacteriol 2002;
Rev Lett 1986;56(9):930–3. 184(14):3992–4002.
 K. Sudesh et al. / Polymer Degradation and Stability 83 (2004) 281–287 287

[21] Pötter M, Madkour MH, Mayer F, Steinbüchel A. Regulation of among short-chain-length polyhydroxyalkanoic acids producing
phasin expression and polyhydroxyalkanoate (PHA) granule bacteria, is a DNA-binding protein and represses Paracoccus
formation in Ralstonia eutropha H16. Microbiology 2002;148(Pt. denitrificans phaP expression in vitro. FEMS Microbiol Lett
8):2413–26. 2001;200(1):9–15.
[22] York GM, Stubbe J, Sinskey AJ. The Ralstonia eutropha PhaR [26] Sudesh K, Gan Z, Doi Y. (Unpublished results).
protein couples synthesis of the PhaP phasin to the presence of [27] Sudesh K, Gan Z, Matsumoto K, Doi Y. Direct observation of
polyhydroxybutyrate in cells and promotes polyhydroxybutyrate polyhydroxyalkanoate chains by atomic force microscopy.
production. J Bacteriol 2002;184:59–66. Ultramicroscopy 2002;91(1–4):157–64.
[23] Iwata T, Doi Y, Kasuya K, Inoue Y. Visualization of enzymatic [28] Vesenka J, Manne S, Giberson R, Marsh T, Henderson E. Col-
degradation of poly[(R)-3-hydroxybutyrate] single crystals by an loidal gold particles as an incompressible atomic force micro-
extracellular PHB depolymerase. Macromolecules 1997;30:833–9. scope imaging standard for assessing the compressibility of
[24] Marchessault RH, Coulombe S, Morikawa H, Okamura K, biomolecules. Biophys J 1993;65:992–7.
Revol J-F. Solid-state properties of poly-b-hydroxybutyrate and [29] Lee WK, Iwata T, Abe H, Doi Y. Studies on the enzymatic
its oligomers. Can J Chem 1981;59:38–44. degradation of solution-grown lamellar crystals of poly[(R)-3-
[25] Maehara A, Doi Y, Nishiyama T, Takagi Y, Ueda S, Nakano H, hydroxybutyrate]: defects in crystals. Macromolecules 2000;
Yamane T. PhaR, a protein of unknown function conserved 33(26):9535–41.